Frühzeitige Detektion von Oberflächenzerrüttungen/Premature detection of surface disruption – Deep Learning-based method for classification of damage on ball screw drives

Die Zustandsüberwachung von Anlagen, Maschinen und deren Bauteilen ist eine zentrale Thematik von Industrie 4.0. Unvorhergesehene Ausfälle von Werkzeugmaschinen sind häufig auf den Verschleiß und das daraus resultierende Versagen von Kugelgewindetrieben zurückzuführen. Aufgabe dieser Arbeit ist die frühzeitige Detektion von Oberflächenschäden auf der Kugelgewindetriebspindel mit einem elektromechanischen Kamerasystem in Kombination mit Deep-Learning-basierten Modellen, um entsprechende Wartungsmaßnahmen abzuleiten.   Condition monitoring of plants, machines and their components is a central topic of Industry 4.0. Unforeseeable failures of machine tools are often caused by wear, resulting in failure of ball screws and subsequent surface disruptions. This article describes how image-based monitoring of ball screws by an electronic camera system in combination with deep learning-based models enable the early detection of surface disruptions and to derive appropriate and preventive maintenance measures.

Download Full-text

Integration von Machine Vision in Kugelgewindespindeln*/Integration of machine vision in ball screw drives – Integrated system for condition monitoring of ball screw drives

wt Werkstattstechnik online ◽

10.37544/1436-4980-2019-07-08-95 ◽

2019 ◽

Vol 109 (07-08) ◽

pp. 605-610

Author(s):

T. Schlagenhauf ◽

J. Hillenbrand ◽

B. Klee ◽

J. Fleischer

Keyword(s):

Early Detection ◽

Machine Vision ◽

Condition Monitoring ◽

Machine Tools ◽

Integrated System ◽

Ball Screw ◽

Camera System ◽

Ball Screw Drives

Unvorhergesehene Maschinenausfälle von Werkzeugmaschinen durch natürlichen Verschleiß sind häufig auf den Kugelgewindetrieb zurückzuführen. Für eine frühzeitige Erkennung der auftretenden Schäden, präsentiert dieser Beitrag einen Ansatz für die Überwachung von Spindeln von Kugelgewindetrieben mittels integriertem Kamerasystem. Ziel ist die frühzeitige Detektion von Schäden, die auf der Spindeloberfläche erscheinen, um entsprechende Wartungsmaßnahmen abzuleiten.   Unforeseen failures of machine tools due to wear are often caused by ball screws. To allow for an early detection of damage, this article presents an approach for monitoring ball screw spindles using an integrated camera system. The aim is to detect initial defects that appear on the spindle surface and derive appropriate maintenance measures.

Download Full-text

Integrated Diagnostic and Preload Control for Ball Screw Drives by Means of Self-Sensing Actuators

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.769.271 ◽

2013 ◽

Vol 769 ◽

pp. 271-277 ◽

Cited By ~ 1

Author(s):

Christopher Ehrmann ◽

Stefan Herder

Keyword(s):

Signal Processing ◽

Economic Integration ◽

Machine Tools ◽

Processing System ◽

Ball Screw ◽

Measured Signal ◽

Mechatronic Systems ◽

Signal Processing System ◽

High Flexibility ◽

Ball Screw Drives

Piezoelectric ceramics can be used as sensors, as well as actors. The concept of a self- sensing-actuator tries to use both modes of operation in one device, allowing the economic integration of mechatronic systems. Possible fields of application are ball screws of machine tools, where wear-induced degradation of the preload can be compensated. Furthermore, the signal processing part of such a system can be used to gather information related to the condition of the ball screw. Both excitation signal generation and filtering of the measured signal have to offer high flexibility and signal fidelity. In this article the concept of a power amplifier and its corresponding signal processing system are presented.

Download Full-text

Deep learning enables high-throughput early detection and classification of bacterial colonies using time-lapse coherent imaging (Conference Presentation)

Optics and Biophotonics in Low-Resource Settings VI ◽

10.1117/12.2547399 ◽

2020 ◽

Author(s):

Hongda Wang ◽

Hatice C. Koydemir ◽

Yunzhe Qiu ◽

Bijie Bai ◽

Yibo Zhang ◽

...

Keyword(s):

Deep Learning ◽

Early Detection ◽

High Throughput ◽

Time Lapse ◽

Coherent Imaging

Download Full-text

Review of the State of the Art of Deep Learning for Plant Diseases: A Broad Analysis and Discussion

Plants ◽

10.3390/plants9101302 ◽

2020 ◽

Vol 9 (10) ◽

pp. 1302 ◽

Cited By ~ 1

Author(s):

Reem Ibrahim Hasan ◽

Suhaila Mohd Yusuf ◽

Laith Alzubaidi

Keyword(s):

Machine Learning ◽

Deep Learning ◽

Early Detection ◽

Feature Fusion ◽

State Of The Art ◽

Vital Role ◽

Considerable Improvement ◽

Plant Diseases ◽

Recent Emergence

Deep learning (DL) represents the golden era in the machine learning (ML) domain, and it has gradually become the leading approach in many fields. It is currently playing a vital role in the early detection and classification of plant diseases. The use of ML techniques in this field is viewed as having brought considerable improvement in cultivation productivity sectors, particularly with the recent emergence of DL, which seems to have increased accuracy levels. Recently, many DL architectures have been implemented accompanying visualisation techniques that are essential for determining symptoms and classifying plant diseases. This review investigates and analyses the most recent methods, developed over three years leading up to 2020, for training, augmentation, feature fusion and extraction, recognising and counting crops, and detecting plant diseases, including how these methods can be harnessed to feed deep classifiers and their effects on classifier accuracy.

Download Full-text

Industrie-4.0-Nachrüstkit für Werkzeugmaschinen/Modular solution for condition-based maintenance and process monitoring – Industry 4.0 retrofitting kit for machine tools

wt Werkstattstechnik online ◽

10.37544/1436-4980-2020-07-08-35 ◽

2020 ◽

Vol 110 (07-08) ◽

pp. 491-495 ◽

Cited By ~ 1

Author(s):

David Barton ◽

Reinhard Stamm ◽

Sebastian Mergler ◽

Cedric Bardenhagen ◽

Jürgen Fleischer

Keyword(s):

Process Monitoring ◽

Machine Tools ◽

Industry 4.0 ◽

Economic Benefit ◽

Industrie 4.0 ◽

High Potential ◽

Condition Based Maintenance ◽

Machining Processes ◽

Software Modules ◽

Roll Out

Industrie 4.0 verspricht ein hohes wirtschaftliches Potenzial für produzierende Unternehmen. Allerdings wird dieses in bestehenden Werkzeugmaschinen bisher nur wenig ausgeschöpft. Um das Ausrollen von Funktionen für die zustandsorientierte Instandhaltung und die Überwachung des Bearbeitungsprozesses zu ermöglichen, wurde ein modulares Nachrüstkit entwickelt. Mit dem Kit können Maschinen individuell um Hardware- und Softwarebausteine erweitert werden.   Industry 4.0 offers manufacturers a high potential for economic benefit. However, this potential is only rarely exploited in existing machine tools. To enable the roll-out of functions for condition-based maintenance and monitoring of machining processes, a modular retrofitting kit has been developed. This kit allows machines to be individually upgraded with hardware and software modules.

Download Full-text

Classification of retinal images with deep learning for early detection of diabetic retinopathy disease

2018 26th Signal Processing and Communications Applications Conference (SIU) ◽

10.1109/siu.2018.8404369 ◽

2018 ◽

Cited By ~ 3

Author(s):

Nursel Yalcin ◽

Seyfullah Alver ◽

Necla Uluhatun

Keyword(s):

Diabetic Retinopathy ◽

Deep Learning ◽

Early Detection ◽

Retinal Images

Download Full-text

Automated Detection and Classification of Oral Lesions Using Deep Learning for Early Detection of Oral Cancer

IEEE Access ◽

10.1109/access.2020.3010180 ◽

2020 ◽

Vol 8 ◽

pp. 132677-132693 ◽

Cited By ~ 2

Author(s):

Roshan Alex Welikala ◽

Paolo Remagnino ◽

Jian Han Lim ◽

Chee Seng Chan ◽

Senthilmani Rajendran ◽

...

Keyword(s):

Deep Learning ◽

Oral Cancer ◽

Early Detection ◽

Automated Detection ◽

Oral Lesions

Download Full-text

The Printed Machine Tool for Micro Machining

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.1018.433 ◽

2014 ◽

Vol 1018 ◽

pp. 433-440 ◽

Cited By ~ 1

Author(s):

Christoph Batke ◽

Karl Heinz Wurst ◽

Armin Lechler ◽

Alexander Verl

Keyword(s):

Machine Tool ◽

Machine Tools ◽

Ball Screw ◽

Work Piece ◽

Micro Machining ◽

Motion Generation ◽

Alternative Drive ◽

Space Saving ◽

Cost Efficient ◽

Ball Screw Drives

Machine tools for micro machining are so far not adapted to work piece sizes and process forces. They feature hardly any modularity and do not allow reconfiguration in a significant process change. One possibility to adapt the machines is to produce them from plastic or composite materials through generative methods. This “printed” machine is a reconfigurable, monolithic module, in which drives are integrated. By a cooperative motion generation, larger workspaces can be realized while the installation spaces decreases. This gives the possibility to use alternative drive technologies for example piezo-drives. Based on these methods, two small generatively produced machine tools are designed. These machine tools use two different drive principles. The first machine tool is equipped with ball screw drives which are cost efficient and space saving. The second machine tool uses piezo-actuators, which are very dynamic in motion generation. Further has to be examined, which tolerances and rigidities are needed at critical points and which parts can be produced generatively and which in a conventional way.

Download Full-text

Multi- class classification of breast cancer abnormalities using Deep Convolutional Neural Network (CNN)

PLoS ONE ◽

10.1371/journal.pone.0256500 ◽

2021 ◽

Vol 16 (8) ◽

pp. e0256500

Author(s):

Maleika Heenaye-Mamode Khan ◽

Nazmeen Boodoo-Jahangeer ◽

Wasiimah Dullull ◽

Shaista Nathire ◽

Xiaohong Gao ◽

...

Keyword(s):

Breast Cancer ◽

Neural Network ◽

Deep Learning ◽

Early Detection ◽

Convolutional Neural Network ◽

Learning Model ◽

Learning Rate ◽

Deep Convolutional Neural Network ◽

Deep Learning Model

The real cause of breast cancer is very challenging to determine and therefore early detection of the disease is necessary for reducing the death rate due to risks of breast cancer. Early detection of cancer boosts increasing the survival chance up to 8%. Primarily, breast images emanating from mammograms, X-Rays or MRI are analyzed by radiologists to detect abnormalities. However, even experienced radiologists face problems in identifying features like micro-calcifications, lumps and masses, leading to high false positive and high false negative. Recent advancement in image processing and deep learning create some hopes in devising more enhanced applications that can be used for the early detection of breast cancer. In this work, we have developed a Deep Convolutional Neural Network (CNN) to segment and classify the various types of breast abnormalities, such as calcifications, masses, asymmetry and carcinomas, unlike existing research work, which mainly classified the cancer into benign and malignant, leading to improved disease management. Firstly, a transfer learning was carried out on our dataset using the pre-trained model ResNet50. Along similar lines, we have developed an enhanced deep learning model, in which learning rate is considered as one of the most important attributes while training the neural network. The learning rate is set adaptively in our proposed model based on changes in error curves during the learning process involved. The proposed deep learning model has achieved a performance of 88% in the classification of these four types of breast cancer abnormalities such as, masses, calcifications, carcinomas and asymmetry mammograms.

Download Full-text